Help with building your Speeduino, installing it, getting it to run etc.
#44399
Updates in red.
DeeeeC wrote: Fri Jul 10, 2020 12:58 am [3] Things I want to do:
[f] I will use the stock missing tooth wheel (at flywheel) for the CKP and adjust the code if needed to accept this.
No code changes needed as my car uses a 60-2 crank wheel.
[g] Make sure I still get warning lights on my dash and to use the stock dashboard Tach and Speedo. I will add a display for TunerStudio.
Choosing whether to use a big-as* monitor like Tesla or carefully sculpt a tablet into the stock dash.
Extract information from my stock ECU maps and apply them in TunerStudio to get close to stock before I do real tuning.
I have my map but haven't got around to dissecting it yet. I heard it can be difficult.
#44403
You need to measure the injector voltage and current. The DC-DC converter stage can be tricky if you have high current because you can get voltage sag with the turn-on transient, big caps help but can affect the stability. You may also find that you only need high voltage for the peak and then battery voltage for the hold stage.
#44461
JHolland wrote:You need to measure the injector voltage and current. The DC-DC converter stage can be tricky if you have high current because you can get voltage sag with the turn-on transient, big caps help but can affect the stability. You may also find that you only need high voltage for the peak and then battery voltage for the hold stage.
This is where I'm a bit stuck because my car is not starting so I can't just hook up a scope to see what it's like. I'm gonna have to do some bench testing unless I find some info. I found some 'scope traces for similar injectors, it's an interesting form, with a pre-charge, massive spike, high hold, low hold and then off. The stock ecu and spares are all dead - hence the predicament.

I'm thinking, put a spare injector in a jar of acetone and try out a few voltages with some well-insulated wires outside with a bucket of water for safety. If I have to resort to this. Bosch part number 46805546 or 0261 500 013.

I found a motherload of oscilloscope traces here: https://www.forum.alfaholicy.org/156/61 ... anguage=en which I need to read through.
#46513
Many thanks to Andrey and JHolland. Your posts have been very useful. I had a feeling 200V was too high and the body of the injectors looked more like a solenoid - this saved a lot of effort.

The events of the pandemic put me under a huge load of pressure (like everyone). I diverted a lot of personal and work resources to help make face shields and masks in a charity effort. I really super burned out on those efforts, and health took a nose-dive. My clients ended my work contracts and I've just had to divert all efforts to personal survival to be honest. So I'm sorry I've postponed my work on this until better days. I also started a business which made some great products but more than 10,000 potential customers have effectively told me they just can't afford anything non-essential. I can't blame them.

For now I hope this thread helps someone. I hope I can come back to it someday. You all rock!
andrey wrote: Sun Oct 18, 2020 2:07 am Bosch 46805546 looks like a solenoid (not piezo) GDI injector, the ECU case also looks same as 2006 2.0T Bosch MED9.1 ECU.

Those solenoid injectors need "only" 65v - rusEFI has a working board for this (which is really just a cheaper and smaller alternative to NXP dev board).

https://youtu.be/Ha29INY53Sk
andrey wrote: Sun Oct 18, 2020 2:18 am These two pics are pretty cool:

Image

Image
JHolland wrote: Wed Oct 21, 2020 8:34 am There is information on driving high voltage injectors here:

http://publications.lib.chalmers.se/rec ... 219127.pdf

I was one of the hardware designers on the the ECU discussed.
#57017
Hi All,

an update from me. I'm feeling like picking this up again. I spent the last 7 days just thinking about it and fixing a fuel leak and some minor maintenance first. Beware of fake R9 fuel hose that is rated for ethanol enhanced fuels that are in all fuel now (of a small percentage) in the UK. The hose I bought from Continental worked fine, just two small bits from another supplier failed.

Ok, so my concept at this time to make GDI work with Speeduino is quite experimental. I wanted to piggyback so that my stock ECU handles the DBW, Wideband O2 and basically everything except for injectors and ignition. My big concern is I wanted to keep my ABS, Airbags, Traction control, etc etc all working.

The way fuel is managed in my car is the tank pump has a pressure regulator at 4.5 bar and the mechanical fuel pump takes it to 50 bar upwards. There is an electronic high pressure regulator that keeps it at 50 bar at all loads.

Designing an injector driver from scratch is a big effort. It's been mentioned that rusEFI has a way of doing it.

My idea is to totally remove the need for a custom injector driver. Your ideas and feedback would be very welcome and needed.

I am thinking to bypass the high pressure mechanical pump entirely and run off the 4.5 bar pressure from the tank pump. I would do this by blocking off two unneeded (now) ports in the common fuel rail with threaded stoppers and copper gaskets. My dad who is an engineer can make those for me. Speeduino firmware would never know that there had been a 50 bar pump. A lot of cars that run Speeduino might only have 3 or 4 bar to the injectors. This is assuming my injectors can make a nice enough mist and flow enough at a lower fuel pressure.

Then I could use Speeduino as-is. I'd piggy back for everything else and run on Alfa-N. I found the wire in my MAF that gives a temperature reading. The stock IAC works as far as I can tell - but that can be handled by Speeduino also.

Is this idea crazy? If the injectors don't work, I could just get a set of 12V injectors I guess - they are easy to get in dimensions that fit my ports.

I wonder if I could still pass emissions? If I set up the map to stick to stoichiometric, there are two cats to clean up whatever is left. This is to Euro 3 spec. They test exhaust with a probe at idle for a few minutes then at 2000 rpm. My tester did it at 4000 rpm, not sure if that was a favour or not.

Let me know your thoughts. My car is covered in spider webs, even though I washed it two weeks ago.

There are a lot of GDI cars that just end up being scrapped. I rebuilt this engine 3000 miles ago and made a custom polished intake and the sound is music to me, so I don't feel like giving up. I think the previous owners balanced the bottom end because it is that smooth. I'm also planning on the works to make some NA power, such as headers, cams, Water Methanol Injection, and possibly ITBs as my intake was made to be upgraded to this. That's another story though.
#57020
this would probably not work.

Without high fuel pressure and more importantly without 60 volts and peak and hold you would not get DI injectors to flow predicably.

You can convert your vehicle to port injection by drilling manifold and installing legacy port injectors and at that point it's no longer "Direct Injection Project - Let me be the first!" but "let's revert DI engine into MPI"

You can use BMW bosch 7506280 GDI driver or similar Lexus or Hyuindai driver to get you DI injectors happy, but...

but you also need to control electronic relief valve to keep target pressure since without high pressure you would run out of duty cycle while revving engine under load.

open question if target pressure could be same or if it should be variable based on RPM/Load
Vr Conditioner Noise when cranking

New version 202305 don't fix the issue. Now after[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Hello, I bring news!! Let me tell you that after […]

Honda J30 running

Okay, we now need to talk about violence and backg[…]

Still can't find what you're looking for?